Spatial, temporal and relational patterns in respondent-driven sampling: evidence from a social network study of rural drug users

Evaluation of respondent-driven sampling in seven studies of people who use drugs from rural populations: findings from the Rural Opioid Initiative

BACKGROUND

Accurate prevalence estimates of drug use and its harms are important to characterize burden and develop interventions to reduce negative health outcomes and disparities. Lack of a sampling frame for marginalized/stigmatized populations, including persons who use drugs (PWUD) in rural settings, makes this challenging. Respondent-driven sampling (RDS) is frequently used to recruit PWUD. However, the validity of RDS-generated population-level prevalence estimates relies on assumptions that should be evaluated.

METHODS

RDS was used to recruit PWUD across seven Rural Opioid Initiative studies between 2018-2020. To evaluate RDS assumptions, we computed recruitment homophily and design effects, generated convergence and bottleneck plots, and tested for recruitment and degree differences. We compared sample proportions with three RDS-adjusted estimators (two variations of RDS-I and RDS-II) for five variables of interest (past 30-day use of heroin, fentanyl, and methamphetamine; past 6-month homelessness; and being positive for hepatitis C virus (HCV) antibody) using linear regression with robust confidence intervals. We compared regression estimates for the associations between HCV positive antibody status and (a) heroin use, (b) fentanyl use, and (c) age using RDS-1 and RDS-II probability weights and no weights using logistic and modified Poisson regression and random-effects meta-analyses.

RESULTS

Among 2,842 PWUD, median age was 34 years and 43% were female. Most participants (54%) reported opioids as their drug of choice, however regional differences were present (e.g., methamphetamine range: 4-52%). Many recruitment chains were not long enough to achieve sample equilibrium. Recruitment homophily was present for some variables. Differences with respect to recruitment and degree varied across studies. Prevalence estimates varied only slightly with different RDS weighting approaches, most confidence intervals overlapped. Variations in measures of association varied little based on weighting approach.

CONCLUSIONS

RDS was a useful recruitment tool for PWUD in rural settings. However, several violations of key RDS assumptions were observed which slightly impacts estimation of proportion although not associations.

Opioid Overdoses and Take-Home Naloxone Interventions: Ethnographic Evidence for Individual-Level Barriers to Treatment of Opioid Use Disorders in Rural Appalachia

Introduction: Research indicates that take-home naloxone (THN) is saving lives across rural Appalachia, but whether it also results in treatment for opioid use disorders (OUDs) remains unclear. This study involves a detailed qualitative analysis of interviews with 16 individuals who had overdosed on opioids 61 times to understand why a THN intervention does not routinely lead to OUD treatment. Methods: This study builds upon a one-year (2018) qualitative study on community responses to opioid overdose fatalities in four adjacent rural counties in Western Pennsylvania. Using a semi-structured interview guide, 16 individuals who had experienced one or more overdoses were interviewed. Using NVivo, the transcribed audio-recorded interviews were coded, and a thematic analysis of the coded text was conducted. Findings: Findings reveal that of the 29 overdoses that included a THN intervention, only eight resulted in treatment. The analysis derives five individual-level barriers to treatment: (1) opioid dependence, (2) denial/readiness, (3) opioid withdrawal fears, (4) incarceration concerns, and (5) stigma and shame. These barriers impeded treatment, even though all the interviewees knew of treatment programs, how to access them, and in some cases had undergone treatment previously. Discussion and Conclusion: findings indicate that there is evidence that the five barriers make entering treatment after a THN intervention challenging and seemingly insurmountable at times. Recommendations based on the findings include increasing efforts to reduce stigma of OUDs in the community, including self-stigma resulting from misusing opioids, increasing informational efforts about Good Samaritan Laws, and increasing familiarity with medication-assisted treatments for OUDS.

Extent and implications of omitted ties on network measures in a longitudinal social network survey of people who use drugs

BACKGROUND

Previous research has revealed under-reporting of personal network members (i.e., alters) in studies involving people who use drugs (PWUD). This analysis (1) characterizes relationships that were more likely to be omitted but later recalled with prompting and (2) identifies network structural characteristics most impacted by these omissions among a sample of PWUD in rural Appalachian Kentucky, an epicenter of the opioid epidemic.

METHODS

Data were collected through longitudinal assessments as part of the Social Networks Among Appalachian People (SNAP) study (2008-2017). Study participants completed interviewer-administered questionnaires that collected social network data via free-listing at baseline and six-month intervals. At visit 5, after free-listing, interviewers prompted participants with the names of previously reported alters. We used modified Poisson regression with generalized estimating equations to identify individual- and relationship-level characteristics associated with an alter being reported only after prompting. We examined the impact of including vs. excluding relationships reported after prompting on local and global sociometric network measures (i.e., betweenness centrality, bridging, density, mean degree, transitivity, cliques, and 2-cores).

RESULTS

Relationships reported only after prompting were more likely to be immediate family (Adjusted Prevalence Ratio [APR]:1.29; 95% Confidence Interval [CI]: 1.03-1.63) and less likely to involve sex (APR:0.54; 95% CI: 0.43-0.67). Considerable differences were observed for participant positional rankings of betweenness centrality and bridging, and differences in network density and average degree pre- and post-prompting were statistically significant.

CONCLUSION

Longitudinal network studies that aim to assess transmission dynamics, information diffusion, or peer influence should consider the effects of omitted relationships.

Peer-based Retention Of people who Use Drugs in Rural Research (PROUD-R2): a multisite, randomised, 12-month trial to compare efficacy of standard versus peer-based approaches to retain rural people who use drugs in research

INTRODUCTION

Rural communities bear a disproportionate share of the opioid and methamphetamine use disorder epidemics. Yet, rural people who use drugs (PWUD) are rarely included in trials testing new drug use prevention and treatment strategies. Numerous barriers impede rural PWUD trial engagement and advancing research methods to better retain rural PWUD in clinical trials is needed. This paper describes the Peer-based Retention Of people who Use Drugs in Rural Research (PROUD-R2) study protocol to test the effectiveness of a peer-driven intervention to improve study retention among rural PWUD.

METHODS AND ANALYSIS

The PROUD-R2 study is being implemented in 21 rural counties in three states (Kentucky, Ohio and Oregon). People who are 18 years or older, reside in the study area and either used opioids or injected any drug to get high in the past 30 days are eligible for study inclusion. Participants are allocated in a 1:1 ratio to two arms, stratified by site to assure balance at each geographical location. The trial compares the effectiveness of two retention strategies. Participants randomised to the control arm provide detailed contact information and receive standard retention outreach by study staff (ie, contacts for locator information updates, appointment reminders). Participants randomised to the intervention arm are asked to recruit a 'study buddy' in addition to receiving standard retention outreach. Study buddies are invited to participate in a video training and instructed to remind their intervention participant of follow-up appointments and encourage retention. Assessments are completed by intervention, control and study buddy participants at 6 and 12 months after enrolment.

ETHICS AND DISSEMINATION

The protocol was approved by a central Institutional Review Board (University of Utah). Results of the study will be disseminated in academic conferences and peer-reviewed journals, online and print media, and in meetings with community stakeholders.

TRIAL REGISTRATION NUMBER

NCT03885024.

Longitudinal trends in nonmedical prescription opioid use in a cohort of rural Appalachian people who use drugs

Rural Appalachia remains an epicenter of the prescription opioid epidemic. In 2008, a cohort study was undertaken to examine longitudinal trends in nonmedical prescription opioid use (NMPOU). Eight waves of data (2008-2020) from the Social Networks among Appalachian People (SNAP) cohort were utilized for the current analysis. Only those who reported recent (past 6-month) NMPOU at baseline are included (n = 498, 99%). Mixed-effects logistic regression was used to model factors associated with NMPOU over time. Recent NMPOU declined significantly over the past decade (p < .001). However, 54.1% of participants still engaged in NMPOU at their most recent follow-up. Receipt of benefits for a physical or mental disability (adjusted odds ratio [aOR]: 3.11, 95% Confidence Interval [CI]: 1.98, 4.90) and self-described poor health status (aOR: 3.67, 95% CI: 1.61, 8.37) were both associated with NMPOU. All treatment modalities (methadone maintenance, residential, outpatient counseling) tested in the model, with the notable exception of detoxification, were associated with significantly lower odds of NMPOU. Although significant declines in prescription opioid misuse were observed in the cohort, more than half of all participants were engaged in NMPOU more than a decade after entering the study. Substance use disorder (SUD) treatment (excluding detoxification) was shown associated with reduced odds of continued NMPOU; therefore, increasing access to evidence-based treatments should be a priority in rural areas affected by the ongoing opioid epidemic.

Opioid-related overdose deaths among African Americans: Implications for research, practice and policy

Opioid-related overdose deaths among African Americans have only recently received national attention despite evidence of increase in death rates among this population spanning the past decade. Numerous authors have highlighted how the 'opioid epidemic' has largely been portrayed as a problem mostly affecting White America. The purpose of this commentary is to provide a synthesis spotlighting the unique structural and cultural considerations involved in research, practice and policy related to opioid use and treatment for opioid use disorders among African Americans. The commentary concludes with considerations for future research and practice intended to reduce deaths among this group.

Reassessing geographic bottlenecks in a respondent-driven sampling based multicity study in Brazil

This study analyzes the spatial dynamics of drug users' recruitment chains in the context of a respondent-driven sampling (RDS) study in the city of Recife, Brazil. The purpose is to understand the geographic bottlenecks, influenced by social geography, which have been a major challenge for RDS-based studies. Temporo-spatial analysis was used. Sequential maps depicted the dynamics of the recruiting process, considering neighborhood of residence and/or places of drug use. Poisson regression was fitted to model the recruiting rate by neighborhood of residence and/or places of drug use, and the different neighborhoods' demographics. The distance between neighborhood of residence and/or places of drug use and the assessment center was negatively associated with recruitment. There was a positive association between the proportion of the population living in informal settings and the recruiting rate per neighborhood of residence and/or places of drug use. Recruitment chains depend on the social geography and demographics of the population. Studies should incorporate seeds from as many neighborhoods as possible, and more than one assessment center should be utilized.

Novel Recruitment Methods for Research Among Young Adults in Rural Areas Who Use Opioids: Cookouts, Coupons, and Community-Based Staff

OBJECTIVES

Rural communities in the United States are increasingly becoming epicenters of substance use and related harms. However, best practices for recruiting rural people who use drugs (PWUD) for epidemiologic research are unknown, because such strategies were developed in cities. This study explores the feasibility of web- and community-based strategies to recruit rural, young adult PWUD into epidemiologic research.

MATERIALS AND METHODS

We recruited PWUD from rural Kentucky to participate in a web-based survey about opioid use using web-based peer referral and community-based strategies, including cookouts, flyers, street outreach, and invitations to PWUD enrolled in a concurrent substance use study. Staff members labeled recruitment materials with unique codes to enable tracking. We assessed eligibility and fraud through online eligibility screening and a fraud detection algorithm, respectively. Eligibility criteria included being aged 18-35, recently using opioids to get high, and residing in the study area.

RESULTS

Recruitment yielded 410 complete screening entries, of which 234 were eligible and 151 provided complete, nonfraudulent surveys (ie, surveys that passed a fraud-detection algorithm designed to identify duplicate, nonlocal, and/or bot-generated entries). Cookouts and subsequent web-based peer referrals accounted for the highest proportion of screening entries (37.1%, n = 152), but only 29.6% (n = 45) of entries from cookouts and subsequent web-based peer referrals resulted in eligible, nonfraudulent surveys. Recruitment and subsequent web-based peer referral from the concurrent study yielded the second most screening entries (27.8%, n = 114), 77.2% (n = 88) of which resulted in valid surveys. Other recruitment strategies combined to yield 35.1% (n = 144) of screening entries and 11.9% (n = 18) of valid surveys.

CONCLUSIONS

Web-based methods need to be complemented by context-tailored, street-outreach activities to recruit rural PWUD.

Estimation and correction of bias in network simulations based on respondent-driven sampling data

Respondent-driven sampling (RDS) is widely used for collecting data on hard-to-reach populations, including information about the structure of the networks connecting the individuals. Characterizing network features can be important for designing and evaluating health programs, particularly those that involve infectious disease transmission. While the validity of population proportions estimated from RDS-based datasets has been well studied, little is known about potential biases in inference about network structure from RDS. We developed a mathematical and statistical platform to simulate network structures with exponential random graph models, and to mimic the data generation mechanisms produced by RDS. We used this framework to characterize biases in three important network statistics – density/mean degree, homophily, and transitivity. Generalized linear models were used to predict the network statistics of the original network from the network statistics of the sample network and observable sample design features. We found that RDS may introduce significant biases in the estimation of density/mean degree and transitivity, and may exaggerate homophily when preferential recruitment occurs. Adjustments to network-generating statistics derived from the prediction models could substantially improve validity of simulated networks in terms of density, and could reduce bias in replicating mean degree, homophily, and transitivity from the original network.

Privacy, Confidentiality, and Safety Considerations for Conducting Geographic Momentary Assessment Studies Among Persons Who Use Drugs and Men Who Have Sex with Men

Geographic momentary assessments (GMA) collect real-time behavioral data in one's natural environment using a smartphone and could potentially increase the ecological validity of behavioral data. Several studies have evaluated the feasibility and acceptability of GMA among persons who use drugs (PWUD) and men who have sex with men (MSM), but fewer have discussed privacy, confidentiality, and safety concerns, particularly when illegal or stigmatized behavioral data were collected. This study explores perceptions regarding privacy, confidentiality, and safety of GMA research among PWUD and MSM recruited in three different settings (rural Appalachia, a mid-sized city in the South, and a mid-Atlantic city). Between November 2014 and April 2017, we recruited 35 PWUD from rural Appalachian Kentucky (N = 20) and Baltimore, Maryland (N = 15) and 20 MSM from Lexington, Kentucky to complete semi-structured qualitative interviews. Through thematic analyses, we identified and compared privacy, confidentiality, and safety concerns by demographic characteristics, risk behaviors, and setting. Privacy, confidentiality, and safety concerns varied by setting, age, smartphone ownership, use of illegal drugs, and history of drug-related arrests. Among those who used drugs, participants reported concerns with being tracked and burden associated with carrying and safeguarding study phones and responding to survey prompts. Privacy and confidentiality concerns were noted in each setting, but tracking concerns were greatest among Baltimore participants and led many to feel that they (or others) would be unwilling to participate or comply with study procedures. While locations considered to be sensitive varied by setting, participants in all settings said they would take measures to prevent sensitive information from being collected (i.e., intentionally disable devices, leave phones at home, alter response times). Privacy, confidentiality, and safety concerns may limit the accuracy of risk location information, study compliance, and participation. As concerns were often greatest among those engaging in illegal behaviors and with the highest risk behaviors, selection bias and non-response bias could negatively influence the representativeness and validity of study findings.

Peer influence of injection drug use cessation among dyads in rural eastern Kentucky

BACKGROUND

This analysis aims to assess whether injection drug use cessation among peers predicts injection drug use cessation among individuals and explores whether this association varies by relationship type and strength.

METHODS

Data were collected through baseline and 6-month assessments for the Social Networks among Appalachian People study (2008-2011). Interviewer-administered surveys collected sociodemographic and drug use behaviors (past 6 months and lifetime). Participants also listed sex, drug use, and social support partners (past 6 months). Listed names were cross-referenced with survey participants to identify relationships between study participants. The analytic sample was further restricted to include only those relationship pairs where both individuals reported a history of injection drug use at baseline (n = 244 unique individuals and 746 dyads). We fit a generalized estimating equations logistic regression model to (1) assess the relationship between peer injection cessation and individual injection cessation and (2) determine whether the strength of this association differs by relationship-level variables (i.e., relationship role, relationship type, relationship duration, frequency of interaction, residential proximity).

RESULTS

Overall, those with a network member who ceased injection drug use were more likely to stop injecting over the following 6-month period (AOR=1.65). The magnitude of this association was greater for social support partners (AOR=2.95), family members (AOR=3.56), those with whom the participant interacted at least daily (AOR=2.17), and those who the participant knew longer (AOR=2.09). Further, among family members, the effect size was greater when relationships were further restricted to immediate family members (AOR=5.35).

CONCLUSION

Our findings suggest that in this rural community, closer, more supportive relationships, may be more influential for modeling injection cessation; however, relationship-types were not mutually exclusive so differences in effect size across strata may not be statistically significant. In this setting, social support through the recovery process (including cessation attempts with peers) may increase likelihood of injection cessation.

Suboptimal HIV Pre-exposure Prophylaxis Awareness and Willingness to Use Among Women Who Use Drugs in the United States: A Systematic Review and Meta-analysis

In the United States (U.S.), more than 12 million women reported illicit drug use in the past month. Drug use has been linked to increased risk for HIV, but little is known about the uptake of HIV pre-exposure prophylaxis (PrEP) to prevent HIV among women who use drugs (WWUD). Following the PRISMA guideline, we conducted a multi-database literature search to assess engagement along the PrEP care continuum among WWUD in the U.S. Seven studies with a total of 755 women were included in the review: 370 (49%) Black, 126 (16.7%) Hispanics, and 259 (34.3%) Whites. Employing random-effect models, data indicate 20.6% (95% CI 8.7%, 32.4%) of WWUD were aware of PrEP, and 60.2% (95% CI 52.2%, 68.2%) of those aware were also willing to use PrEP. Notwithstanding study limitations, our findings suggest there may be potential to increase PrEP uptake among WWUD, but efforts must first concentrate on improving PrEP awareness among this population.

Motivation to Change and Treatment Participation Among Syringe Service Program Utilizers in Rural Kentucky

PURPOSE

Kentucky experiences a disproportionate burden of substance use disorder (SUD), particularly in rural areas of the state. Multiple factors increase vulnerability to SUD and limit access to services in rural communities. However, the recent implementation and expansion of syringe service programs (SSPs) in rural Kentucky may provide a leverage point to reach at-risk people who inject drugs (PWID).

METHODS

Data were collected as part of an ongoing NIDA-funded study designed to examine uptake of SSPs among PWID in Appalachian Kentucky. Using Respondent Driven Sampling (RDS), the study enrolled a sample of 186 PWID SSP attenders across 3 rural Appalachian Kentucky counties and conducted face-to-face interviews regarding health behaviors, injecting practices, SSP utilization, and treatment services. Using logistic regression analyses, we examined consistent SSP use, as well as importance and confidence to reduce substance use as predictors of current treatment participation.

FINDINGS

For the prior 6 months, 44.6% of the sample reported consistent SSP use. Consistent use of SSPs was associated with treatment participation in the unadjusted logistic regression models. Significant predictors of treatment participation in the adjusted model included high confidence to reduce substance use, and not reporting primary methamphetamine injection.

CONCLUSIONS

Rurally located SSPs may play an important role in supporting confidence and motivation to change substance use behaviors among PWID impacted by SUD. SSPs may be critical venues for integration and expansion of prevention, health promotion, and treatment linkage services for this underserved population.

Network-Based Research on Rural Opioid Use: an Overview of Methods and Lessons Learned

PURPOSE OF REVIEW

The purpose of this paper is to provide a thorough overview of methods used for recruitment, network data collection, and network data management in a network-based study of rural people who use drugs (PWUD) and to offer methodological recommendations for future research on rural drug use.

RECENT FINDINGS

The Social Networks among Appalachian People (SNAP) study recruited a cohort of 503 rural PWUD via respondent-driven sampling (RDS) and has retained more than 80% of eligible participants over 7-9 years. SNAP has yielded important methodological insights, including that (1) RDS referral was non-random and disproportionately involved kin and (2) interviewer-administered questionnaires were successful in eliciting accurate name and age information about network members. The SNAP experience suggests that RDS was a successful recruitment strategy for rural PWUD and questionnaires administered by community-based interviewers in the context of a Certificate of Confidentiality could elicit detailed data on PWUD risk networks.

Physicians' Decision-making When Implementing Buprenorphine With New Patients: Conjoint Analyses of Data From a Cohort of Current Prescribers

OBJECTIVES

Few studies have considered how providers make decisions to prescribe buprenorphine to new patients with opioid use disorder. This study examined the relative importance of patients' clinical, financial, and social characteristics on physicians' decision-making related to willingness to prescribe buprenorphine to new patients and the number of weeks of medication that they are willing to initially prescribe after induction.

METHODS

A national sample of 1174 current prescribers was surveyed. Respondents rated willingness to prescribe on a 0 to 10 scale and indicated the number of weeks of medication (ranging from none to >4 weeks) for 20 hypothetical patients. Conjoint analysis estimated relative importance scores and part-worth utilities for these 2 outcome ratings.

RESULTS

The mean rating for willingness to prescribe was 5.52 (SD 2.47), indicating a moderate willingness to implement buprenorphine treatment. The mean prescription length was 2.06 (SD 1.34), which corresponds to 1 week of medication. For both ratings, the largest importance scores were for other risky substance use, method of payment, and spousal involvement in treatment. Illicit benzodiazepine use, having Medicaid insurance to pay for the office visit, and having an opioid-using spouse were negatively associated with these outcome ratings, whereas a history of no risky alcohol or benzodiazepine use, cash payment, and having an abstinent spouse were positively associated with both ratings.

CONCLUSIONS

Reticence to prescribe to individuals using an illicit benzodiazepine and individuals with a drug-using spouse aligns with practice guidelines. However, reluctance to prescribe to patients with Medicaid may hamper efforts to expand access to treatment.

Social Networks of Substance-Using Populations: Key Issues and Promising New Approaches for HIV

PURPOSE OF REVIEW

This paper presents recent literature on substance using networks and HIV, highlighting renewed and emerging themes in the field. The goal is to draw attention to research that holds considerable promise for advancing our understanding of the role of networks in shaping behaviors, while also providing critical information for the development of interventions, programs, and policies to reduce HIV and other drug-related harms.

RECENT FINDINGS

Recent research advances our understanding of networks and HIV, including among understudied populations, and provides new insight into how risk environments shape the networks and health of substance-using populations. In particular, the integration of network approaches with molecular epidemiology, research on space and place, and intervention methods provides exciting new avenues of investigation. Continued advances in network research are critical to supporting the health and rights of substance-using populations and ensuring the development of high-impact HIV programs and policies.

Using Network and Spatial Data to Better Target Overdose Prevention Strategies in Rural Appalachia

This analysis uses network and spatial data to identify optimal individuals to target with overdose prevention interventions in rural Appalachia. Five hundred and three rural persons who use drugs were recruited to participate in the Social Networks among Appalachian People Study (2008-2010). Interviewer-administered surveys collected information on demographic characteristics, risk behaviors (including overdose history), network members, and residential addresses. We restricted the sample to individuals with at least one confirmed relationship to another study participant (N = 463). Using dyadic analyses (N = 1428 relationships), we identified relationship-level correlates of relationships with network members who have previously overdosed. We then examined individual- and network-level factors associated with (1) having at least one first-degree alter (i.e., network member) with a prior overdose and (2) each additional network member with a prior overdose (N = 463 study participants). Overall, 28% of the sample had previously overdosed and 57% were one-degree away from someone who previously overdosed. Relationships with those who had overdosed were characterized by closer residential proximity. Those with at least one network member who previously overdosed were more geographically central and occupied more central network positions. Further, the number of network members with an overdose history increased with decreasing distance to the town center, increasing network centrality, and prior enrollment in an alcohol detox program. Because fatal overdoses can be prevented through bystander intervention, these findings suggest that strategies that target more central individuals (both geographically and based on their network positions) and those who have previously enrolled in alcohol detox programs with overdose prevention training and naloxone may optimize intervention reach and have the potential to curb overdose fatalities in this region.

Silence Surrounding Hepatitis C Status in Risk Relationships Among Rural People Who Use Drugs

Hepatitis C virus (HCV) has reached epidemic proportions in rural Central Appalachia in recent years. We sought to identify demographic, behavioral, and interpersonal characteristics associated with HCV serostatus disclosure among high risk people who use drugs (PWUD) in Appalachian Kentucky. HCV antibody-positive participants (n = 243), drawn from the fifth follow-up assessment of a longitudinal study of rural PWUD, completed interviewer-administered questionnaires eliciting demographic and interpersonal characteristics, risk behaviors, and information about HCV disclosure. We assessed correlates of HCV disclosure using gender-stratified multivariate logistic regression. Participants reported having disclosed their HCV-positive status to a current sex partner (44.0%), family member (35.8%), close friend (9.5%), or past sex partner (6.6%). Of those reporting current (n = 72) or past (n = 215) injection drug use (IDU), only 2.8% disclosed to current and 0.9% disclosed to past IDU partners, respectively. Female participants were more likely than male participants to disclose to current sex partners and family member(s). In multivariate analyses, adjusting for time since testing HCV positive, older age and lifetime history of drug treatment were associated with decreased odds of HCV disclosure among females, while only lifetime history of drug treatment was associated with decreased odds of HCV disclosure among males. In summary, the almost complete absence of disclosure to current or former injection drug use partners was concerning. However, most participants (69.1%) reported disclosing their HCV status to at least one of their social referents, suggesting that family members, partners, and friends of people living with HCV could play a critical role in encouraging uptake of treatment. Although further research is warranted, it is clear that interventions are needed to encourage HCV disclosure among those most at risk of transmitting HCV.

Overlooked Threats to Respondent Driven Sampling Estimators: Peer Recruitment Reality, Degree Measures, and Random Selection Assumption

Intensive sociometric network data were collected from a typical respondent driven sample (RDS) of 528 people who inject drugs residing in Hartford, Connecticut in 2012-2013. This rich dataset enabled us to analyze a large number of unobserved network nodes and ties for the purpose of assessing common assumptions underlying RDS estimators. Results show that several assumptions central to RDS estimators, such as random selection, enrollment probability proportional to degree, and recruitment occurring over recruiter's network ties, were violated. These problems stem from an overly simplistic conceptualization of peer recruitment processes and dynamics. We found nearly half of participants were recruited via coupon redistribution on the street. Non-uniform patterns occurred in multiple recruitment stages related to both recruiter behavior (choosing and reaching alters, passing coupons, etc.) and recruit behavior (accepting/rejecting coupons, failing to enter study, passing coupons to others). Some factors associated with these patterns were also associated with HIV risk.

Use of social network analysis and global sensitivity and uncertainty analyses to better understand an influenza outbreak

In the summer of 2014, an influenza A(H3N2) outbreak occurred in Yichang city, Hubei province, China. A retrospective study was conducted to collect and interpret hospital and epidemiological data on it using social network analysis and global sensitivity and uncertainty analyses. Results for degree (χ²=17.6619, P<0.0001) and betweenness(χ²=21.4186, P<0.0001) centrality suggested that the selection of sampling objects were different between traditional epidemiological methods and newer statistical approaches. Clique and network diagrams demonstrated that the outbreak actually consisted of two independent transmission networks. Sensitivity analysis showed that the contact coefficient (k) was the most important factor in the dynamic model. Using uncertainty analysis, we were able to better understand the properties and variations over space and time on the outbreak. We concluded that use of newer approaches were significantly more efficient for managing and controlling infectious diseases outbreaks, as well as saving time and public health resources, and could be widely applied on similar local outbreaks.

Identification of Homophily and Preferential Recruitment in Respondent-Driven Sampling

Respondent-driven sampling (RDS) is a link-tracing procedure used in epidemiologic research on hidden or hard-to-reach populations in which subjects recruit others via their social networks. Estimates from RDS studies may have poor statistical properties due to statistical dependence in sampled subjects' traits. Two distinct mechanisms account for dependence in an RDS study: homophily, the tendency for individuals to share social ties with others exhibiting similar characteristics, and preferential recruitment, in which recruiters do not recruit uniformly at random from their network alters. The different effects of network homophily and preferential recruitment in RDS studies have been a source of confusion and controversy in methodological and empirical research in epidemiology. In this work, we gave formal definitions of homophily and preferential recruitment and showed that neither is identified in typical RDS studies. We derived nonparametric identification regions for homophily and preferential recruitment and showed that these parameters were not identified unless the network took a degenerate form. The results indicated that claims of homophily or recruitment bias measured from empirical RDS studies may not be credible. We applied our identification results to a study involving both a network census and RDS on a population of injection drug users in Hartford, Connecticut (2012-2013).

A rural/urban comparison of privacy and confidentiality concerns associated with providing sensitive location information in epidemiologic research involving persons who use drugs

BACKGROUND

Analyses that link contextual factors with individual-level data can improve our understanding of the "risk environment"; however, the accuracy of information provided by participants about locations where illegal/stigmatized behaviors occur may be influenced by privacy/confidentiality concerns that may vary by setting and/or data collection approach.

METHODS

We recruited thirty-five persons who use drugs from a rural Appalachian town and a Mid-Atlantic city to participate in in-depth interviews. Through thematic analyses, we identified and compared privacy/confidentiality concerns associated with two survey methods that (1) collect self-reported addresses/cross-streets and (2) use an interactive web-based map to find/confirm locations in rural and urban settings.

RESULTS

Concerns differed more by setting than between methods. For example, (1) rural participants valued interviewer rapport and protections provided by the Certificate of Confidentiality more; (2) locations considered to be sensitive differed in rural (i.e., others' homes) and urban (i.e., where drugs were used) settings; and (3) urban participants were more likely to view providing cross-streets as an acceptable alternative to providing exact addresses for sensitive locations and to prefer the web-based map approach.

CONCLUSION

Rural-urban differences in privacy/confidentiality concerns reflect contextual differences (i.e., where drugs are used/purchased, population density, and prior drug-related arrests). Strategies to alleviate concerns include: (1) obtain a Certificate of Confidentiality, (2) collect geographic data at the scale necessary for proposed analyses, and (3) permit participants to provide intersections/landmarks in close proximity to actual locations rather than exact addresses or to skip questions where providing an intersection/landmark would not obfuscate the actual address.

Examining the Social Context of Injection Drug Use: Social Proximity to Persons Who Inject Drugs Versus Geographic Proximity to Persons Who Inject Drugs

In this analysis, we used social network and spatial data to examine associations between people's drug injection status and their social and/or spatial proximity to others who injected drugs. We recruited 503 rural Kentucky residents who used drugs to participate in the Social Networks among Appalachian People (SNAP) Study (2008-2010). Interviewer-administered surveys collected information on recent (past 6 months) sex, drug-use, and social-support network members (n = 897 ties). Using network simulations, we determined a threshold for the association between social proximity to others who injected drugs and recent injection status ("socially proximal" was defined by a shortest path ≤2). We defined "geographically proximal" as the median road-network distance between pairs of individuals who both injected drugs (≤7 miles (≤11.2 km)). Logistic regression was used to determine the independent and joint associations between the number of socially and/or geographically proximal injecting peers and a person's injection status. After adjustment, the odds of recent injection increased by 0.4% for each injecting peer who was geographically proximal but not socially proximal, 12% for each geographically and socially proximal injecting peer, and 22% for each injecting peer who was socially proximal but not geographically proximal. When implementing network-based interventions which promote cessation of injection drug use, investigators should consider collecting sociometric network data to examine whether the intervention diffuses through the network and whether there are additive or threshold effects.

Rural and urban injection drug use in Puerto Rico: Network implications for human immunodeficiency virus and hepatitis C virus infection

Understanding the short- and long-term transmission dynamics of blood-borne illnesses in network contexts represents an important public health priority for people who inject drugs and the general population that surrounds them. The purpose of this article is to compare the risk networks of urban and rural people who inject drugs in Puerto Rico. In the current study, network characteristics are drawn from the sampling "trees" used to recruit participants to the study. We found that injection frequency is the only factor significantly related to clustering behavior among both urban and rural people who inject drugs.

Accuracy of name and age data provided about network members in a social network study of people who use drugs: implications for constructing sociometric networks

PURPOSE

Network analysis has become increasingly popular in epidemiologic research, but the accuracy of data key to constructing risk networks is largely unknown. Using network data from people who use drugs (PWUDs), the study examined how accurately PWUD reported their network members' (i.e., alters') names and ages.

METHODS

Data were collected from 2008 to 2010 from 503 PWUD residing in rural Appalachia. Network ties (n = 897) involved recent (past 6 months) sex, drug cousage, and/or social support. Participants provided alters' names, ages, and relationship-level characteristics; these data were cross-referenced to that of other participants to identify participant-participant relationships and to determine the accuracy of reported ages (years) and names (binary).

RESULTS

Participants gave alters' exact names and ages within two years in 75% and 79% of relationships, respectively. Accurate name was more common in relationships that were reciprocally reported and those involving social support and male alters. Age was more accurate in reciprocal ties and those characterized by kinship, sexual partnership, recruitment referral, and financial support, and less accurate for ties with older alters.

CONCLUSIONS

Most participants reported alters' characteristics accurately, and name accuracy was not significantly different in relationships involving drug-related and/or sexual behavior compared to those not involving these behaviors.

Comparison of Respondent Driven Sampling Estimators to Determine HIV Prevalence and Population Characteristics among Men Who Have Sex with Men in Moscow, Russia

Analytically distinct estimators have been proposed for the calculation of population-based estimates derived from respondent-driven sampling (RDS), yet there have been few comparisons of the inferences from these estimators using empirical data. We compared estimates produced by unweighted analysis used to calculate sample proportions and by three available estimators that are used to calculate population proportions, RDS-I, RDS-II (Volz-Heckathorn), and Gile’s RDS-SS. Data were derived from a cross-sectional, RDS study of men who have sex with men (MSM) conducted from October 2010 to April 2013 in Moscow, Russia (N = 1,376, recruitment depth: 31 waves). Analyses investigated the influence of key parameters: recruitment depth, homophily, and network size on sample and population estimates. Variability in results produced by the estimators and recruitment depth were statistically compared using the coefficient of variation (CV). Sample proportions had the least variability across different recruitment depths, compared to the RDS estimators. Population estimates tended to differ at lower recruitment depth but were approximately equal after reaching sampling equilibrium, highlighting the importance of sampling to greater recruitment depth. All estimators incorporate inverse probability weighting using self-reported network size, explaining the similarities in across population estimates and the difference of these estimates relative to sample proportions. Current biases and limitations associated with RDS estimators are discussed.

Individual and Network Correlates of Antisocial Personality Disorder Among Rural Nonmedical Prescription Opioid Users

PURPOSE

Examination of the association of antisocial personality disorder (ASPD) with substance use and HIV risk behaviors within the social networks of rural people who use drugs.

METHODS

Interviewer-administered questionnaires were used to assess substance use, HIV risk behavior, and social network characteristics of drug users (n = 503) living in rural Appalachia. The MINI International Psychiatric Interview was used to determine whether participants met DSM-IV criteria for ASPD and Axis-I psychological comorbidities (eg, major depressive disorder, posttraumatic stress disorder, generalized anxiety disorder). Participants were also tested for herpes simplex 2, hepatitis C, and HIV. Multivariate generalized linear mixed modeling was used to determine the association between ASPD and risk behaviors, substance use, and social network characteristics.

RESULTS

Approximately one-third (31%) of participants met DSM-IV criteria for ASPD. In multivariate analysis, distrust and conflict within an individual's social networks, as well as past 30-day use of heroin and crack, male gender, younger age, lesser education, heterosexual orientation, and comorbid MDD were associated with meeting diagnostic criteria for ASPD.

CONCLUSIONS

Participants meeting criteria for ASPD were more likely to report recent heroin and crack use, which are far less common drugs of abuse in this population in which the predominant drug of abuse is prescription opioids. Greater discord within relationships was also identified among those with ASPD symptomatology. Given the elevated risk for blood-borne infection (eg, HIV) and other negative social and health consequences conferred by this high-risk subgroup, exploration of tailored network-based interventions with mental health assessment is recommended.

Respondent driven sampling of wheelchair users: A lack of traction?

BACKGROUND

Internationally, wheelchair users are an emerging demographic phenomenon, due to their increased prevalence and rapidly increasing life-span. While having significant healthcare implications, basic robust epidemiological information about wheelchair users is often lacking due, in part, to this population's 'hidden' nature. Increasingly popular in epidemiological research, Respondent Driven Sampling (RDS) provides a mechanism for generating unbiased population-based estimates for hard-to-reach populations, overcoming biases inherent within other sampling methods. This paper reports the first published study to employ RDS amongst wheelchair users.

METHODS

Between October 2015 and January 2016, a short, successfully piloted, internet-based national survey was initiated. Twenty seeds from diverse organisations were invited to complete the survey then circulate it to peers within their networks following a well-defined protocol. A predetermined reminder protocol was triggered when seeds or their peers failed to respond. All participants were entered into a draw for an iPad.

RESULTS

Overall, 19 people participated (nine women); 12 initial seeds, followed by seven second-wave participants arising from four seeds . Completion time for the survey ranged between 7 and 36 minutes. Despite repeated reminders, no further people were recruited.

DISCUSSION

While New Zealand wheelchair user numbers are unknown, an estimated 14% of people have physical impairments that limited mobility. The 19 respondents generated from adopting the RDS methodology here thus represents a negligible fraction of wheelchair users in New Zealand, and an insufficient number to ensure equilibrium required for unbiased analyses. While successful in other hard-to-reach populations, applying RDS methodology to wheelchair users requires further consideration. Formative research exploring areas of network characteristics, acceptability of RDS, appropriate incentive options, and seed selection amongst wheelchair users is needed.

HIV Prevalence Among People Who Inject Drugs in Greater Kuala Lumpur Recruited Using Respondent-Driven Sampling

The HIV epidemic in Malaysia is concentrated among people who inject drugs (PWID). Accurate estimates of HIV prevalence are critical for developing appropriate treatment and prevention interventions for PWID in Malaysia. In 2010, 461 PWID were recruited using respondent-driven sampling in Greater Kuala Lumpur, Malaysia. Participants completed rapid HIV testing and behavioral assessments. Estimates of HIV prevalence were computed for each of the three recruitment sites and the overall sample. HIV prevalence was 15.8 % (95 % CI 12.5-19.2 %) overall but varied widely by location: 37.0 % (28.6-45.4 %) in Kampung Baru, 10.3 % (5.0-15.6 %) in Kajang, and 6.3 % (3.0-9.5 %) in Shah Alam. Recruitment extended to locations far from initial interview sites but was concentrated around discrete geographic regions. We document the high prevalence of HIV among PWID in Greater Kuala Lumpur. Sustained support for community surveillance and HIV prevention interventions is needed to stem the HIV epidemic among PWID in Malaysia.

Assessing the geographic coverage and spatial clustering of illicit drug users recruited through respondent-driven sampling in New York City

We assess the geographic coverage and spatial clustering of drug users recruited through respondent-driven sampling (RDS) and discuss the potential for biased RDS prevalence estimates. Illicit drug users aged 18-40 were recruited through RDS (N = 401) and targeted street outreach (TSO) (N = 210) in New York City. Using the Google Maps API™, we calculated travel distances and times using public transportation between each participant's recruitment location and the study office and between RDS recruiter-recruit pairs. We used K function analysis to evaluate and compare spatial clustering of (1) RDS vs. TSO respondents and (2) RDS seeds vs. RDS peer recruits. All participant recruitment locations clustered around the study office; however, RDS participants were significantly more likely to be recruited within walking distance of the study office than TSO participants. The TSO sample was also less spatially clustered than the RDS sample, which likely reflects (1) the van's ability to increase the sample's geographic heterogeneity and (2) that more TSO than RDS participants were enrolled on the van. Among RDS participants, individuals recruited spatially proximal peers, geographic coverage did not increase as recruitment waves progressed, and peer recruits were not less spatially clustered than seeds. Using a mobile van to recruit participants had a greater impact on the geographic coverage and spatial dependence of the TSO than the RDS sample. Future studies should consider and evaluate the impact of the recruitment approach on the geographic/spatial representativeness of the sample and how spatial biases, including the preferential recruitment of proximal peers, could impact the precision and accuracy of estimates.

HIV vaccine acceptability among high-risk drug users in Appalachia: a cross-sectional study

Background

A vaccine could substantially impact the HIV epidemic, but inadequate uptake is a serious concern. Unfortunately, people who use drugs, particularly those residing in rural communities, have been underrepresented in previous research on HIV vaccine acceptability. This study examined HIV vaccine acceptability among high-risk drug users in a rural community in the United States.

Methods

Interviewer-administered questionnaires included questions about risk behavior and attitudes toward HIV vaccination from 433 HIV-negative drug users (76% with history of injection) enrolled in a cohort study in Central Appalachia. HIV vaccine acceptability was measured on a 4-point Likert scale. Generalized linear mixed models were used to determine correlates to self-report of being “very likely” to receive a 90% effective HIV vaccine (i.e. “maximum vaccine acceptability”, or MVA). Adjusted odds ratios (AORs) and corresponding 95% confidence intervals (CIs) are reported.

Results

Most (91%) reported that they would accept a preventive HIV vaccine, but concerns about cost, dosing, transportation constraints, vaccine-induced seropositivity, and confidentiality were expressed. Cash incentives, oral-administration, and peer/partner encouragement were anticipated facilitators of uptake. In multivariate analysis, men were significantly less likely to report MVA (AOR: 0.33, CI: 0.21 – 0.52). MVA was more common among participants who believed that they were susceptible to HIV (AOR: 2.31, CI: 1.28 – 4.07), that an HIV vaccine would benefit them (AOR: 2.80, CI: 1.70 – 4.64), and who had positive experiential attitudes toward HIV vaccination (AOR: 1.85, CI: 1.08 – 3.17). MVA was also more common among participants who believed that others would encourage them to get vaccinated and anticipated that their behavior would be influenced by others' encouragement (AOR: 1.81, 95% 1.09 – 3.01).

Conclusions

To our knowledge, this study was among the first to explore and provide evidence for feasibility of HIV vaccination in a rural, high-risk population in the United States. This study provides preliminary evidence that gender-specific targeting in vaccine promotion may be necessary to promoting vaccine uptake in this setting, particularly among men. The data also underscore the importance of addressing perceived risks and benefits, social norms, and logistical constraints in efforts to achieve widespread vaccine coverage in this high-risk population.